Editorial
Monitoring and improving CO2 concentration in classroom based on AIoT technology
@ARTICLE{10.4108/tsoe.10749, author={Oanh Tran Thi Hoang and Hung Nguyen Xuan and Ngoc Nguyen Quang and Phu Nguyen Ngoc}, title={Monitoring and improving CO2 concentration in classroom based on AIoT technology}, journal={EAI Endorsed Transactions on Transportation Systems and Ocean Engineering}, volume={1}, number={1}, publisher={EAI}, journal_a={TSOE}, year={2025}, month={12}, keywords={AIoT, LR, MLP, predicted CO2 concentration, ThingSpeak}, doi={10.4108/tsoe.10749} }- Oanh Tran Thi Hoang
Hung Nguyen Xuan
Ngoc Nguyen Quang
Phu Nguyen Ngoc
Year: 2025
Monitoring and improving CO2 concentration in classroom based on AIoT technology
TSOE
EAI
DOI: 10.4108/tsoe.10749
Abstract
The paper presents the monitoring and improvement of CO₂ concentration in the classroom through an AIoT system including Raspberry Pi 3 connected to the MH - Z19B CO₂ gas sensor, DHT11 temperature - humidity sensor and Raspberry Pi camera. The collected data including temperature, humidity, current CO₂ gas and the number of people in a 70 m2 classroom are continuously sent to Thing Speak for analysis and forecasting CO₂ concentration in the classroom using the Multi-Layer Perceptron (MLP) model and linear regression model (LR). The forecasted CO₂ gas concentration results help to give early warnings and control the fan system in the classroom when the predicted CO₂ concentration is greater than 1000 ppm to maintain safe air quality in the classroom, improve concentration and health of learners. The system was tested in real classroom conditions, showing 95.77% accuracy with LR and 98.21% with MLP in predicting CO₂ concentration. This research contributes to improving classroom air quality, contributing to protecting health and improving students' learning efficiency.
